In the beginning, I wasn't really thinking too much about how to figure out if something was pareidolia or not, but rather I was approaching this subject from the point of view of a sort of demonstration of known pareidolia, with the intent to define the parameters of pareidolia that way.
But Tom got me thinking about the subject: how can we tell which is which.
In an earlier post on the Faces topic, jrich commented that maybe if we ask a sufficiently large number of people to see if their pattern recognition detected it, we might learn something that way. Neil immediately attacked this line of thinking as a kind of "collective subjectivism". But, now that I consider this notion of how pareidolic images are "in the eye of the beholder", it starts to seem like what jrich suggested has some merit.
Here's the quote I'm refering to:
<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by neilderosa</i><blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><br /><i>originally posted by jrich</i><br />So one way to objectify the process would be to measure the rate of recognition among a sufficiently large population sample and compare it against a control group exposed to known pareidolia in imagery of similar character.--jrich<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">I disagree fundamentally with this logic. In fine, it amounts to a kind of “collective subjectivism equals objectivism; otherwise known as, ”fifty million Frenchmen can’t be wrong.” --Neil<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"> Now what jrich is suggesting here makes perfect sense to me. If the population was sufficiently large, the control of them responding to known pareidolia with say a question: what is this? might later be compared to both non-pareidolia and Neil's images. I'm fairly sure that this might reveal the truth. Obviously, to protect the integrity of the test Mars wouldn't be mentioned, and there would have to be parameters defined for how much leeway there is in each answer.
Or in short, while it might not be true that "fifty million Frenchmen have to be right", it may be true that 30 Frenchmen will be consistent in their ability to tell "what it is" when viewing known pareidolia, non-pareidolia, and "The Mars Faces."
Another quote from jrich:
<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">If one is going to take the position that ANY "art" pattern that anyone might see is a legitimate candidate for artificiality, then one must also concede that at least some of these candidates are actually pareidolia. So it seems to me it might be useful to have a method of differentiating them as objectively as possible. Obviously, to do this one cannot rely on the existing visual evidence itself since that is precisely what is in doubt. So we must examine the response of the human pattern recognition system to the visual stimuli to judge the validity of the patterns. Whether this can actually be done with any validity is another question entirely.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
In a way, this idea goes back to my "Cat" demonstration early in the thread. But I wouldn't jump to conclusions. I could easily lose this debate based on this type of study. For one thing, I can imagine a case where the question "what is this?" elicits a wide and varied response in the non-pareidolia images, if they were picked correctly.
rd
