Slashdot Mirror
Computer Faces Human Psychological Test
A reader writes: "
The test known as the MMPI-2 (Minnesota Multiphasic Personality Inventory) is going to be administered to a computer-based personality. The test subject is GAC - Generic Artifical Consciousness." This is the first time I've heard of GAC - but this test is quite different from the Turing Test.
> maniac...
I can't reveal that either, when I figure it out about a client . . .
hawk, esq.
I'm sorry, the idea just gives me the creeps.
DNA just wants to be free...
will they give this "personality" the Scientology personality test? Is this AI a promising pre-Clear(TM)? Can it be audited all the way to OT(TM)? In college, one of our passtimes was to take and retake the Scientologists' tests using composite personalities (gather in threes, answer every third).
G)enerator of A)rrogant C)laims.
Well, maybe, but as for me, I hope it does well on this since I can't help but think that if it does, the best way to really test its intelligence would be to follow up the MMPI with some MCSE exams.
If it's truly intelligent, there's two possible reactions: A) GAC goes on a killing spree as it attempts to rid the world of its evil human parasitic infestation; or, B) GAC is reduced to a pile of drooling mush, in which case, we'll know to be a bit more careful with the fragile intellect of GAC 2.
ABSURDITY, n.: A statement or belief manifestly inconsistent with one's own opinion.
AI is much less in the mainstream spotlights than it was thirty years ago but is still progressing. Maybe you could even say it is progressing much better away from the hype. I've helped work on and used AI's that were real enough to be a little scary. Virtual pets are fairly common these days and human-level intelligence/emotions are almost within reach so possibly a psychological test is appropiate. I think building strong psuedo-emotion systems into AI's is obviously an important and oft forgotten part of creating realistic systems that act like living creatures. Luckily Eliza is far from the bleeding edge these days. :)
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
I thought it stood for Great American Country. On MY cable service it does.
--
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
Who let Hemos post without hitting the preview button and checking his URLs?
Does anyone else see a double standard where we are viewed as the only stupid ones?
*grin*
will the test subject know how to close italic tags?
BilldaCat
> The MMPI is quite comprehensive. It is designed so that it is effectively impossible to hide anything.
Interesting. I've heard a several times over the years, from completely independent sources, that a number of states had banned the MMPI because it was prone to giving false diagnoses. (Never bothered to investigate the claims, though. Can I "Ask Slashdot"?)
An associated claim is that it was standardized by benching it on institutionalized people, so that it essentially assumes that everyone has problems; it only decides which category you fit best.
--
Sheesh, evil *and* a jerk. -- Jade
I have submitted a large number of mindpixels to GAC (and verified ten times as many). I have yet to see it answer questions better than 50% of the time. In my more suspicious moments, I wonder if it's all a big hoax, and that there's just a random number generator underneath it. Sometimes I wonder if it will ever work. There seems to be a lot of natural language recognition work yet to be done.
However, if the engine ever does work, it will know quite a lot. Even if most of it is about pr0n.
--
--
E_NOSIG
When are we gonna make a computer program that can take the SAT's?
:)
And of course, the follow-up question: When will that program be ported to my College-Board approved TI-85?
But what is that actual reality? Knowing that you're subject to illusions and delusions, how do you verify that your observations correspond to external objective reality? (If such a thing even exists, but that's a whole other topic...)
Humans are so good at being delusional that Dick's test is useless - when we stop believing in it, we often just stop seeing it! And vice-versa - when we believe in it, we see it whether it's there or not. Our political and social structures are rife with examples of this. (And let's not even start on religion...)
Tom Swiss | the infamous tms | http://www.infamous.net/
Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood
Tom Swiss | the infamous tms | http://www.infamous.net/
Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood
GAC is a failing AI effort which serves no purpose except for Chris McKinstry to draw attention to himself.
The site began by offering users shares of the "Corpus" worth approximately 25 cents per question you submitted. Eventually, the running total of the "money" you had earned disappeared. Chris claimed that the software that generated that page was broken, but it has never come back. There was also a "referral program" along the lines of alladvantage, where you could get partial credit for the "Mindpixels" submitted by people you referred. That quietly went away. Asking Chris where the "money" went gets you ignored, and the fact is he didn't really promise money, just shares in whatever money this "Corpus" eventually makes. I suppose he pulled out at a good time, when he realized the total would be somewhere around $0.
He promised a rating system to get rid of the dumb or repetitive questions and reward the good ones - this never appeared.
A while later, he got the idea that he could create HAL with the same deficient software that ran GAC, just by telling users to answer questions as they thought the computer HAL would answer. Most people saw that this made no sense whatsoever. The massive funding Chris was expecting for HAL never appeared, and he wrote a bitter news item about it.
So now he's managed to get GAC signed up for this Turing-Test-ish-thing. GAC still appears to answer randomly on the simplest true/false questions, so I doubt it will do spectacularly. It's just another way to draw attention to Chris.
--
Win dain a lotica, en vai tu ri silota
Several threads have already recycled the myth that you "can't fool" the MMPI because it repeats questions and checks your answers for consistency. According to an NPR interview I heard several years ago with the people who designed the MMPI, however, this simply isn't true.
It _is_ true that the MMPI repeats questions (at least it used to be true; I assume it still does), but the explanation has nothing to do with something sophisticated like checking to see whether you answer the same questions the same way. Instead, when the original designers of the test first wrote it, they discovered that they didn't have enough questions for the grading machine they were using. Their machines (using punch cards I think) required that a certain number of answers be marked or they would spit out the answer sheet, but the test designers were short a few questions, and didn't have any questions they particularly wanted to add. So they simply interspersed a random handful of questions throughout the test in order to get the grading machine to accept the answer sheets.
Later on, when the grading equipment improved and no longer required a certain number of questions, they retained the extra questions, because by that time the MMPI had already become a widely-used standard, and there was no reason to change it. But the repeated questions never entered into the grade, and were never checked for consistency. The scientists admitted to NPR that they knew (now, years after the fact) that having the repeat questions made people uncomfortable, but this was no part of the design of the test, and never entered into the grading of the test.
Still, whenever the MMPI comes up in conversation, someone claims that you "can't fool" the MMPI because it repeats questions and checks for consistency.
Anything worth doing is worth doing badly -- G.K. Chesterton
That said, it's fucked up. One of my roommates had to take it once (aside: we moved him out the next semester:), and he let us read it. There are 567 "yes/no" questions, and the first one is "I enjoy reading sports car magazines." Others include such obviously gender biased items as "I sometimes wear women's clothing." Basically, my point is, it's a very US-centric test designed to "catch" people who think differently than our definition of "normal." I suspect that the AI device wouldn't even have a coherent answer to many of the questions, since it's never lived in American culture, and doesn't have a childhood to be analysed as "traumatic," etc etc.
I can't see anything useful coming from this except perhaps a false reassurance that "this system won't go nuts like HAL-9000."
---
Once a test has been administered several years and answers of several thousand people have been recorded with the actual information about how these people are outside the test (based on observation or known facts) you can start matching new answer patterns with old ones.
An alarm should go off if your pattern is very similiar to a known rapist/murderer.
Interesting you mention Dick. I just reread Martian Timeslip and, while I enjoyed it as much as I did as a kid, I was struck by his facile conception of schizophrenia. At any rate, a quote from Dick on questions of sanity, madness and reality sheds heat but no light. Dick begs the question of the relationship of reality to normative constructs, as well as the reliability and purity of our observation. In MT, the only sane view, that is corresponding to actual reality, is finally that of the two madmen, the kid and the protagonist.
Certainly, in a society of delusional types, a realist should not be considered mad. Although they almost certainly will be.
The tyranny of the normative is only just as dangerous as that of absolutes. I can make no apology for fools mistaking the two. Such dangers are the cost of doing cultural business. The realist's response is not to dismiss normative judgements out of hand, but to modify the norm. Such a view also cannot preclude other voices without damage to itself, even as their otherness can only be established with reference to that view.
illegitimii non ingravare
self-fulfilling prophecies are hard to escape. if someone labels you as having the symptoms of 'syndrome XYZ', more than likely you'll end up with syndrome XYZ.
The REAL sam_at_caveman_dot_org is user ID 13833.
I'm wondering how GAC will score on the gender section of the MMPI. I hope they got a good cross-section on their data enterers/validators... I know I've had CS courses before where I was the only woman there.
-= rei =-
"This may be presumptuous..." "That's my favorite kind of 'This'."
Laf!
;)
I got this question on Mindpixel:
"I have often wished I were a girl. (Or if you are a girl) I have never been sorry that I am a girl."
That's an MMPI question! Hehee, methinks GAK is studying up....
-= rei =-
"This may be presumptuous..." "That's my favorite kind of 'This'."
I think it's much, much closer than most people expect...
What we're actially doing with GAC is high dimensional tomography. At the moment we're just mapping the 'truth' of each of these high-d points, but soon, we will also measure the 'emotionality' of each point as well. This involves getting people to tell the system how each particular item makes them feel by selecting emotive words from some drop down lists - much like the way 'Quality' is measured now.
See my Slashdot interview for some more links on this emotional thread.
Eliza: Can you elaborate on that?
How can we continue to believe in a just universe and freedom to eat crackers if we have no ale?
I can see it now..
"We don't know who struck first. But we do know that it was us who scorched the skies."
I'll only be freaked out when they actually build an AI computer and it happens to be solar powered.
"The universe seems neither benign nor hostile, merely indifferent." --Carl Sagan
halfway down the page - Some stuff about this 'GAC'
And something about the test itself.
/. is a commercial entity. goto slashdot.com
Does this mean that the first artificial consciense will be a /. fan? ;)
science is a religion
What does it say about the designers of the personality, that they feel they first nered to insure they havn't created a psychopath...?
--CTH
--
--Got Lists? | Top 95 Star Wars Line
Actually, the somatic symptoms are in line with research on depression, which suggests depression is manifested somatically in many cultures.
Very interesting if what you say is true, and GAC is reflecting these profiles.
A lot has been made of AI; this brings up the whole issue of AE (artificial emotion). If an AI is able to exhibit typical emotional profiles based on typical associative/NN models, it suggests AE might not be as far off as some thought.
Your score: somewhat deviant, highly paranoid with delusions of grandeur. However, since this is Slashdot, you're no different from the rest of us.
Hey freaks: now you're ju
Endless arguments over trivial contradictions in books written by ignorant savages to explain thunder in the dark.
Endless arguments over trivial contradictions in books written by ignorant savages to explain thunder in the dark.
@ Mindpixel, this link. You can also "talk" to GAC via this site.
GAC says: Visit our sponsors or donate $2.
--
I think the answer to: Computers, such as GAC, will someday conquer the world and enslave the human race. is:
TRUE
"Your superior intellect is no match for our puny weapons!"
The test won't, however, say that you are a serial killer or homicidal maniac...
That's ok, everyone already knows that a thinking computer is going to start a war against all humans, slaughtering millions and using their skin to clothe humanoid robots to hunt down the remnants of our species like vermin.
Either that or they're all robin williams and haley joel osment
what the hell is a 'junk character', anyway?
Help, help! We're being repressed!
--
--
Mmm... delicious white marbles...
Another problem with having computers calculated to fit their users' personalities would be that it would cause people to separate themselves from society. I'd lock myself in my office and spend hours in conversation with my computer, until my brain is mush and my once-picture-perfect physique resembles a cheese puff. Not that I don't spend hours locked in my office in conversation anyway, but it consists of: "****! Crashed again, you stupid excuse for a toaster oven!" Not exactly scintillating dialogue.
________
"And if the fool, or the pig, are of a different opinion...." -- J.S. Mill
One of the neat things about the MMPI is that it has low face validity. This means, essentially, that it is difficult to determine what any particular question is getting at. The advantage to this is that it is hard to "beat" the test by giving answers that will result in an uncorrect assessment. There are a number of questions designed to check to see if you are trying to appear crazy or too normal. In addition, there are a number of subscales that are combined to provide a psychological profile of the test taker.
Most surprising, however, is the fact that statistical techniques utilizing the MMPI have been shown to be MORE effective than trained clinicians at diagnosing psychological disorders.[1]
Of course, none of this says anything about GAC taking the MMPI.
[1] Meehl, P. E. (1959) A comparison of clinicians with five statistical methods of identifying psychotic MMPI profiles. _Journal of Counseling Psychology,6_, 102-109.
--
Buy Hex-Rated Stuff, fight the DMCA!
G)enerator of A)rrogant C)laims.
For the first time, a standard psychological test used by clinicians worldwide in the evaluation and treatment of adults will be administered to a machine-based artificial personality.
The test is known as the MMPI-2 (Minnesota Multiphasic Personality Inventory).Developed as a specialized psychological test for the measurement of psychopathology, the MMPI has been the preferred psychosocial diagnostic instrument among clinicians for the past 50 years.
Originally published in 1940 by Hathaway and McKinley, MMPI has been implemented in many clinical and non-clinical contexts, including medical, educational, medicolegal and organizational settings. A restandardisation and partial revision of the MMPI resulted in the publication of the MMPI-2 in 1989.
GAC (Generic Artificial Consciousness -- pronounced "Jack") is the artificial personality being developed at the Mindpixel Digital Mind Modeling Project with the collaboration of nearly 40,000 Internet users from more than 200 countries worldwide.
GAC will be evaluated using the MMPI-2 over the next several months to assess its learning of human consensus experience from the Mindpixel project's large and diverse group of users from many different cultures.
The test will be supervised and interpreted by Dr. Robert Epstein, one of the world's leading experts on human and machine behavior.
"Nothing like this has ever been attempted," said Epstein. "We're evaluating thousands of people worldwide as if they were one collective individual."
"We don't know if it is possible to build a normal personality out of millions of little pieces. This experiment will tell us how reasonable the idea is," Epstein added.
In the nearly one year the project has been online, Mindpixel's Internet contributors have made nearly 8 million individual measurements of more than 355,000 individual items of human consensus experience.
The project's organizers hope that they will gain enough information by the time the project's data collection phase is complete (2010) to build a highly accurate statistical model of an average human mind which they hope can be used as a foundation for true artificial consciousness.
One of the world's leading experts on human and machine behavior, Robert Epstein received his doctorate in psychology at Harvard University in 1981. He is Editor-in-Chief of Psychology Today magazine and University Research Professor at United States International University in San Diego.
He is also the founder and Director Emeritus of the Cambridge Center for Behavioral Studies in Massachusetts and Adjunct Professor of Psychology at San Diego State University. He was also the former director of the famed Loebner Prize competition in Artificial Intelligence.
The Mindpixel Digital Mind Modeling Project was launched on July 6, 2000. It is the world's largest Artificial Intelligence effort, with nearly 40,000 contributing members in more than 200 countries.
[Contact: Dr. Robert Epstein, Christopher McKinstry]
02-Jul-2001
use LaTeX? want an online reference manager that
-- john
rodentia [hey, wasn't she spammer?] writes:
"Finally judgements about normativity, while they strike some as distasteful, are really the only way to generalize about psychological phenomena."
Certainly not. Philip K. Dick said "Reality is that which, when you stop believing in it, doesn't go away." One measure of sanity is whether your conception of reality agrees with the actual reality. Certainly, in a society of delusional types, a realist should not be considered mad.
r:
"The idea of universal sanity or a generalized idea of madness are simply absurd and possibly dangerous."
Certainly, they are not as dangerous as a culturally influenced definition of same. Just ask a gay person.
Become a FSF associate member before the low #s are used
The MMPI is quite comprehensive. It is designed so that it is effectively impossible to hide anything. My psych professor said that he took one and (knowing how the test was designed) he tried to cover up that he had a major traumatic event in his life (which he did). The test showed it anyhow.
The test won't, however, say that you are a serial killer or homicidal maniac...
marotti.com
IRTP: I run this project.
You are right, GAC is one heck of a psychological test on its own. However, unlike the MMPI, GAC has not explicitly been tested with both normal and clinical populations, so it would not be good at distinguishing the two.
The whole point of giving the MMPI to GAC now is to see what type of personality you get when you blend together 40,000 personalities. Looking at the data as it comes in (after recovering from the slashdot hit that killed the site for a while), GAC looks a little depressed and slightly sexually deviant. GAC also is showing some strong somatic symptoms (complaints about the body) which is of course quite ironic considering it doesn't have a body. But other than that, GAC appears normal. Quite human.
--------
I think the answer to: Donating $2.00 via PayPal is a good way to contribute to the mindpixel project is:
FALSE
---
Video meliora proboque deteriora sequor - Ovidius
The site is slashdotted, so it's difficult to determine what exactly is going on.
About the beauty of the MMPI: the MMPI is empirically validated, meaning that its scales were developed by selecting items that correlate with an objective criterion. Because items might correlate with an objective criterion (e.g., homicide) without looking like they would, you can get "subtle" scales--scales that are difficult to fake responses to.
But what's the point of giving GAC this?
Say the test predicts the computer is a psychopathic deviate. Does this mean the GAC is not functioning like a normal individual? Does it mean the GAC is psychopathic? Does it mean the average internet user who supplied observations is a psychopathic deviate?
What's ironic is that GAC may actually be a more comprehensive psychological resource than the MMPI. By selecting observations from so many individuals, GAC has tremendous data on what typical thought patterns are.
In fact, GAC could be used as a psychological test itself. This may be why they are giving GAC the MMPI. If GAC is "normal", so to speak, you could use GAC as a test. Even if it's not normal, you could "adjust" GAC to have responses more in line with a "normal" MMPI profile.
Then, you could "test" individuals by having them converse with GAC. The extent to which you "agree" with GAC or not is a gauge of the deviance of your personality. Ideally, it would be just like having a conversation with someone, only the extent to which you get along with GAC reflects on your abnormality.
Interesting...
"Dr. Sbaitso" _WAS_ Eliza. Same algortihm and everything...only named differently so as not to pay any royalties to the copy of "101 BASIC Computer Games" they lifted it from. They stuck in a stupid age check if you tried to talk to the doc about sex...which sucks, because I've a feeling that if my Thunderboard could have helped me through my youthful urges, I wouldn't be the cross dressing deviant I am today.
Hey freaks: now you're ju
From Mindpixel's website:
The Mindpixel Corpus is the world's first and largest database of validated consensus human knowledge. It is an ever improving approximation of the mind of an average Internet user, constructed and owned tens of thousands of people just like you from all over the world, speaking all of the world's languages.
Emphesis added by me.
So, this thing is going to be obsessed with eBay and pr0n.
--
"We don't know if it is possible to build a normal personality out of millions of little pieces. This experiment will tell us how reasonable the idea is," Epstein added.
I predict that the result will be about as useful as Congress.
Aah, change is good. -- Rafiki
Yeah, but it ain't easy. -- Simba
You've gotta love this stealth advertizing for Spielberg's AI!
Refrag
I have a website. It's about Macs.
Didn't we try this 30 years ago? Did it work then?
--
--Got Lists? | Top 95 Star Wars Line
Is fixed here.