Slashdot Mirror
Innocent Adults Are Easy To Convince They Committed a Serious Crime
binarstu (720435) writes "Research recently published [link is to abstract only; full text requires subscription] in Psychological Science quantifies how easy it is to convince innocent, "normal" adults that they committed a crime. The Association for Psychological Science (APS) has posted a nice summary of the research. From the APS summary: "Evidence from some wrongful-conviction cases suggests that suspects can be questioned in ways that lead them to falsely believe in and confess to committing crimes they didn't actually commit. New research provides lab-based evidence for this phenomenon, showing that innocent adult participants can be convinced, over the course of a few hours, that they had perpetrated crimes as serious as assault with a weapon in their teenage years."
Basically, don't talk to the police without a lawyer present. Period. I mean, I'm not going to stonewall a cop that pulls me over for a broken taillight, but if the line of questioning goes any further than what's immediately relevant to said taillight, that's when I shut up. And you can guarantee that I will be videotaping the entire encounter! Cops are under no obligation to tell you the truth about anything; it's up to you to know what your rights are in a given situation and assert them.
It's one of the few things that are quite black and white, either someone committed a crime or he did not.
A plea bargain is not someone saying they partially committed a crime, it is them admitting full guilt to a crime. The law frequently gives a range of punishments, e.g. upper and lower limits for fines or jail time, and the judge can pick something lower if the defendant saves everyone a bunch of time by just admitting it. Some laws also allow judges to remit or change the charges to a lesser crime depending on the circumstances too, especially if part of the plea bargain involves helping them catch other people. It has nothing to do with partly committing a crime, but what amount of punishments and exactly what charges are used. That said, it can pressure innocent people in tough situations to just accept guilt.
I did fail a polygraph even though I was telling the complete truth. It was a full scope polygraph for a Federal agency. The polygraph examiner - a LEO - then interrogated me for a bit, accusing me of various crimes that he thought I was likely to have committed. While I was disappointed in the outcome, I took advantage of the exchange to observe real-life interrogation techniques. I noted the following:
1. Building up my ego. He told me that, with my experience, he thought I'd be an excellent candidate for the job. I think this was supposed to make me like him, to re-establish the rapport he believed was broken when he told me I failed the test.
2. Claiming he wanted to help me. I just had to give him something he could give to the adjudicators, so they could authorize a retest. From my reading, I knew that post-test admissions are automatically disqualifying. This is the oldest trick in the book - the subject is put in a position of distress, and the interrogator offers to help in exchange for information. Naturally, anything that is divulged will be used against the subject later.
3. Claiming anything I told him would be just between me and the agency. They weren't going to share anything with my employer, another Federal entity. This was a lie, of course. Pre-employment polygraph results are provided to other Federal agencies for periodic reinvestigations.
4. Making unrealistic claims about the polygraph itself. The examiner told me that, in every case he has seen, evidence of deception on the polygraph was backed up by facts. From official US government sources, I knew only about half of such cases were substantiated by confession or investigation.
Having been through the wringer, or at least a light version of it, I can see how a person disposed toward trusting authority figures could be manipulated into making a false confession. The interrogator presents himself as an ally and offers to help while minimizing negative consequences of cooperation. Since the interrogator's definition of cooperation is "give me something I can use against you," the compliant person sees substantial benefit and little harm in making a confession, even a false one.
The polygraph, of course, is mainly a prop to elicit confessions. A DNI said as much to Congress a few years ago. And consider this: In the CIA's detention program, probably the most important interrogation program in modern American history, they didn't even bother polygraphing their detainees when they thought they might be lying.
Particularly in drug crimes, prosecutors routinely use extreme penalties to win plea bargains. There is also a penalty for going to trial: people pleading guilty get much lower sentances on average than those found guilty at trial.
I wasn't familiar with the Reid Technique, but once I learned what it was, it struck me as an incredibly unfair and abusive interrogation technique. It's also the technique we see often on a lot of those police investigation television shows: There's a presumption of guilt, all of the questions are loaded. I never knew it had a name, I always called it, "The Asshole Interrogation Technique," because you have to be an asshole to use it.
For those who are interested, the Wikipedia has a short article but The New Yorker has a much more interesting one.
Love sees no species.
Or you've spent 5 years in an illegal prison without valid charge, and subject to torture.
Finally, prosecution says you can go home tomorrow if you plead guilty to this new retrospective crime we just made up.
And there you have the confession and first conviction of a Gitmo detainee! (Five years later, the US Court of Appeals ruled the conviction invalid.)
http://en.wikipedia.org/wiki/D...
It seems like it is - on paper.
In reality that bell curve is actually skewed to the right - due to the limiting effects of low intelligence and test error.
The peak in the middle of the curve is actually a flat line, so the top of the curve is not a single person with a perfect IQ100, but millions of people scoring AROUND IQ100.
Also, due to the built-in unreliability of the test itself, a certain percentage will certainly score less than their actual intelligence.
Due to stress, various environmental and personal issues, even things like time from last meal or how many hours of sleep one had prior to the test.
For all those things there is no normalizing effect which would increase the score - if there were it would be outselling any drug out there.
So, the curve actually leans to the right at the top, but the measured result comes out as symmetrical due to normalization which distorts the representation of the reality, due to the assumption that the only measurement error is in (in)accurately counting the number of correct answers (which is self-normalizing).
Meanwhile, on the far ends of the curve (both left and right) you got what seems like exactly the same number of geniuses (right) and severely retarded (left).
Except that those under a certain level of IQ can't take part in a society at all and must be housed in special institutions - or they die.
So instead of a perfect 50-50 bell curve, it is closer to 40something - average - 50something division.
Where 50+X % are of above average IQ, average IQ is a certain percentage instead of a peak in the middle, some small percentage of people are either babies, senile, comatose or severely retarded people incapable of taking care of themselves and the remaining 40+Y percent are in the below average but intellectually functional bracket.
There is no limit on the above average side of the curve at which said IQ would negatively affect a persons ability to function.
There is such a limit on the below average side, and it is far above 0.
Mit der Dummheit kämpfen Götter selbst vergebens