My pareidolia knows no bounds.

"The observers were told that the letter S (for “superstitious”) was present on 50% of the 20,000 trials, which were equally divided into 40 blocks and completed over a fortnight. No more detail was given regarding the shape of the letter. The image presented on each trial consisted of static bit noise spanning 50 _ 50 pixels (2° _ 2° of visual angle), with a blackpixel density of 50%. No signal was ever presented. The experiment ran on a G4 Macintosh computer using a program written with the Psychophysics Toolbox for Matlab (Brainard, 1997; Pelli, 1997)."

---

You seem to be quite familliar with these test protocols; thanks for answering for Richie. I'm assuming from what I copied above, which I glossed over the first time around, that a "psych" program scrambled white noise 20,000 times, and that was the data from which the test subjects made their guesses. So it seems a person would actually have to take the test himself to make any kind of assessment. And by taking a computer "composite" (or average) of each subject's answers, you come up with "box a", which actually does look like an S because that was what they were searching for, (random shapes that reminded them of an S). I won't attempt to draw any conclusions as to the relevance of this test to what we are doing, but it's interesting.

Actually I could make an inferrence. We are looking for shapes that remind us of faces (human, cartoon, or animal). Some of them might be random shapes and forms, in which case it would be a kind of optical illusion or pareidolia. And some of them could be real (artistic) faces. That's what we are trying to figure out.

Neil

Neil,

The test is not really directly applicable to our discussion here. What is significant though is that it shows that the human brain can be trained to very effectively find patterns in even completely random noise and that the patterns that it finds are at least somewhat influenced by individual biases concerning what such patterns should look like.

One gets the impression that those who favor the artificiality hypothesis of the land art do so in major part because they are unable to concieve of how their perceptions could be mistaken so often. I think this experiment demonstrates that it is possible to see what one hopes or expects to see, to do so many times, and with increasing proficiency in visual imagery where no artificial patterns actually exist.

JR

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by jrich</i>

One gets the impression that those who favor the artificiality hypothesis of the land art do so in major part because they are unable to concieve of how their perceptions could be mistaken so often. I think this experiment demonstrates that it is possible to see what one hopes or expects to see, to do so many times, and with increasing proficiency in visual imagery where no artificial patterns actually exist.

JR
<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

I really do not agree with this assessment. I see very few of the presumed artworks. Only if the proportions of the face or figure are "dead on" do I see it. The rest, which seem to have some level of distortion or exageration, I write off as "cloud figures".

But the chance of an entire face being so "absolutely correct" that one would identify it with some person down the street....

the odds against this being natural are astronomical.

In most of these cases of accurate art, the local terrain has not been used to build the picture. Rather, they seem like charcoal on paper.

Gregg Wilson

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by jrich</i>
<br />It is unknown (unknowable?) how often an S appeared by chance in the random test images, but it would be substantially less than 50%.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

(jrich, you understood this test pretty much exactly as it was intended, and since all the information was posted, I was a little baffled by all the questions.) I would be inclined to think there were NO actual Ss in the slides. What the subjects are doing was "matching their own knowledge of an S with whatever systematic representation they could find of an S", since they were told that there were Ss there. But if we looked at any one of the slides that they said "Yes" on, we might (probably certainly) would disagree. I'm working on getting a little bit more info along these lines.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"> There are images which may be interpreted to contain an S, but there are no images in which an S was intentionally inserted. The "a" images are sort of the summation of the images where the test subjects saw an S. ....... They probably could have repeated the experiment for every letter in the alphabet using the same test subjects and the same test images and gotten similar results for each iteration.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

Yes, exactly. That's the key point. They are currently duplicating the test using faces.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">6- "Box a" therefore represents, for each subject, the information that, on average, distinguished an S from noise IN AN EXPERIMENT WHEN ONLY WHITE NOISE WAS PRESENTED." Again, are WE seeing the same (real) "box a" with only noise in it?<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"><blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">Not sure what you are asking here.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

"Box a" are all Yeses summed, all the Nos summed and subtracted from the summed Yeses.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"> Forgive me if I don't take the experimenter's word for this. Any valid scientific study must be repeatable, so others can do the same experiment and get the same results. This is what JP did in his "Profile" paper in MRB, we could actually do the optical illusion tests and see for ourselves. This is also what we did in our "T or E" paper; we showed the reader how to overcome the light inversion effect-Neil.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"><blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">Until the experiment is repeated, you are correct to be skeptical, though this is a pretty simple experiment for them to have somehow screwed up.-jrich<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

Remember who is doing this. These are scientists at the highest levels of current research, at the University of Glasgow, UK. I think it's safe to say they know the scientific method, like we know how to drive our cars. I would be embarrassed to question that aspect of it. I spent two years working with men like that, and believe me, they are rigorous way beyond what I think is necessary. I'm more of a "shoot from the hip" kind of researcher.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">8- So, I decided to do a little experiment of my own. I cropped out the three Box a from the paper, and applied three simple image processing techniques that I frequently use. <hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"><blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">I'm not sure what the point is either.--jrich<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
I proved to myself, that the same image processing techniques that I might use on a "face", worked to bring out the Ss. Plus, it gave us more information about the clarity of the S, and how it relates to the number of times the respondent saw an S.



rd

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by jrich</i>
<br />I think this experiment demonstrates that it is possible to see what one hopes or expects to see, to do so many times, and with increasing proficiency in visual imagery where no artificial patterns actually exist.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"> Yes, that's it. I relayed to Dr. Schyns a little story about my wife:

"By the way, a couple of years ago, when we first started "anomaly hunting" in the Mars images, my wife, who is not really a math/science kind of person, but more a literature/arts "people person" kind of person, said in answer to my question about what she thought of the faces on Mars: "If you think they are there, you will find them. If you don't, you won't." Isn't that exactly what the S paper shows? I have a whole new respect for her insight." To which Dr. Schyns replied, "Yes, this is the idea that one sees what one knows. Profound insight!"

It gets more complicated from here, though.

rd

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by rderosa</i>
<br />I proved to myself, that the same image processing techniques that I might use on a “face”, worked to bring out the Ss.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">Can you prove that to any of us? Show us a white noise image and a legitimate image processing technique that will bring out a credible "S" in it. I do not expect this to be possible. -|Tom|-