More faces in clouds

In our recent discussion on cryptic images and randomness, it was pngarrison who raised the issue of pareidolia – the human tendency to see images, especially of faces, in what are clearly (apart from God’s sense of humour) random patterns.sheldrake_bbc1

Here’s another example of a cryptic image (actually created by Rupert Sheldrake) just to remind you of what was in mind.

Now forget for a moment the dispute about whether any formal methodology could determine whether the image herein is intentional or fortuitous, like a pattern of leaves on the ground. All our contributors seem to agree that randomness is at least frequently a pattern we haven’t yet noticed. Instead, let’s briefly ask what human facility enables one to discern it.

The philosopher Ludwig Wittgenstein asked this question, in a wider context including what facility gets jokes or understands music. His conclusion was that it is the imagination, which I would think to be fairly uncontroversial. At some point we get the intuition that maybe the blotches at the top frame a couple of human faces, and suddenly we can see two dancers, in their entirety.

In this case, both the context (you already know it’s a psychological test) and the exactness of the correlation lead to a conviction that the image is intentional and not fortuitous. But belief is one thing – the perception of the image itself comes from the power of imagination, which would seem to be the very same faculty involved in pareidolia.

Whether pngarrison’s insomnia heightened his imagination, or just muddled his ability to evaluate the impression of the imaginary patterns he perceived I don’t know. To have had the wholehearted conviction that someone really had arranged faces in the dirt would have required, I guess, greater mental disturbance than a moderate sleep deficit. Presumably we all want to detect patterns that are real rather than imaginary.

But if Wittgenstein is right (and he merits some respect as a thinker) then without an active imagination we could see neither the illusions nor the genuine patterns. Both would be inaccessible to us, and both would appear random. That isn’t a radically new idea either. As Einstein said:

“Imagination is more important than knowledge. For knowledge is limited to all we now know and understand, while imagination embraces the entire world, and all there ever will be to know and understand.”

That accords, of course, with what we know of how Einstein discovered relativity in the first place. What is less known is that that same intuition involved him more than once in what we might call “scientific pareidolia”. Hans Ohanian, examining his various errors, wrote:

Albert often let his intuition overrule flawed proofs and shaky math.

Einstein was undoubtedly not unique either in the incisiveness of his imagination, nor in its sometimes getting the better of his other faculties. But its an interesting discussion whether it’s worse to have a rich imagination, with some “flawed proofs and shaky math”, or to be logically watertight and deficient in imagination. True knowledge appears to require both.

Avatar photo

About Jon Garvey

Training in medicine (which was my career), social psychology and theology. Interests in most things, but especially the science-faith interface. The rest of my time, though, is spent writing, playing and recording music.
This entry was posted in Philosophy, Science. Bookmark the permalink.

7 Responses to More faces in clouds

  1. Avatar photo Merv Bitikofer says:

    Okay, the “we” in “suddenly we can see two dancers” does not, unfortunately, include me.

    Any more helpful hints for somebody lacking in visual imagination? Is it two faces? Or two full human forms? Facing the viewer? Facing each other? And for that matter… I never did catch on to the more difficult one from a post or two back either. Remedial help needed for the imaginatively impaired …

    If anyone really doubts the non-existence of a demarcation between “patterned” or “random”, just try this thought experiment. We see pattern in a picture because we have seen such scenes before. So, limited as we are, we still recognize at least the “macro-aspects” of what we see … a house here, a cloud there, a person over there. But what if you, like omniscient God, had seen the whole universe through all time. Now what we would have called indisputably random: the initial pattern of rain drops hitting a certain square meter of my back patio at the beginning of a certain rain storm, becomes a recognizeable event. Somebody presents you with the recorded square meter, with its pattern of spots to quiz you about whether or not it is random. And with perfect memory you respond: “That is absolutely not random! Why that’s exactly where the initial raindrops landed in that square meter of Merv’s back patio in St. George in that rainstorm of spring of 2014! … I’d recognize that pattern anywhere”; you say, showing off.

    So once it has happened in our universe, it is part of a reality open to any omniscient observers who may exist. What is meaningless noise to us, becomes the equivalent of rainbows and faces to omniscience.

    • Avatar photo Jon Garvey says:


      The other one (now it can be told!) is a cowboy on a horse. I wonder if it was inspired by an ancient joke in which a kid asks his mum what an abstract picture in a gallery is. “It’s supposed to be a cowboy on a horse.” Kid replies “Then why isn’t it?”

      If it’s any comfort I didn’t use this piccy the first time because I couldn’t “see” it till I looked again yesterday. Sheldrake used it in a TV experiment on “morphic resonance” to see if it became easier for people to recognise once some people already had… you’ve disproved his theory, or shown it wears off after several years, or maybe you’re just the exception.

      No doubt there are different degrees, and even types, of imagination leading to differential powers of cryptic figure recognition. But you endorse the point that it ain’t obvious to all. I’d still expect you to beat any computer.

      Your point about patterns and omniscience is good, but perhaps one needs to factor in not just repetition but purpose to relate it to “true” images. If in my omniscience I notice that two monkeys in different labs happened to type the same 100-character string, it’s only significant if I also recognise the quirk I’d designed into monkey neurology that made that more than just coincidence. It’s, I suppose, the equivalent of a grand master noticing some rare opening move in chess and so knowing how to respond.

      PS on pngarrison’s experience perhaps you should keep yourself awake for a few nights and take a look at the cryptic figures then.

    • pngarrison says:

      I think it’s always appropriate on Jon’s blog to refer to a song, so here’s one. Even when we can’t make out an image, it can still be enjoyable. One of my many unattained musical goals is to learn to play this thing.

      Leave It Like It Is

      • Avatar photo Jon Garvey says:

        Great song – and an interesting meditation on how it relates these two blogs and our wider intentions. Thanksfor posting the link. The guitar seems very like John Renbourn’s style: not surprising that it’s a challenge!

  2. pngarrison says:

    I can’t see this one either, Merv, if it makes you feel better, and no, I not going to stay awake voluntarily or take any speed to see if that makes me see it.

    One perhaps interesting thing about my insomniac pareidolia was that I only saw 2 types of things, faces and words. The former is perhaps not surprising since distinguishing faces is an ability that has presumably been around for a very long time and has a lot of survival value, but it is curious that I would see words, something that only emerged a few thousand years ago. I have no idea what that means, except that our brains are very good at learning pattern recognition. (Or all the captcha thingies on the net wouldn’t work.)

    • Avatar photo Jon Garvey says:

      Guys, there’s a crib for this one from this page (not sure if it’s cheating or not!)

      As you say, imagining faces is very understandable – as well as simply having a long time to have acquired the habit, it’s one of the primitives that infants possess almost as soon as their born – the very first step in their socialisation and mental development.

      My speculation re imagining words: writing is the commonest activity in which we have to identify arbitrary symbols accurately – and especially when handwritten the actual shapes vary greatly. As I’m a typical doctor, very few people can see any meaning in my handwriting apart from what they imagine themselves.

  3. pngarrison says:

    I happened to come across a maybe relevant reference. The nice thing about misplacing PNAS references is that when you find them >6 mo later, the article is by then freely available. I haven’t read it all yet, so I’ll postpone any comment.

    The roots of folk biology

Leave a Reply