[EDIT] I just realized that your reply will likely come in the form of a publication and not a blog like this one, so perhaps I should rephrase and say I’d simply like to read it when it becomes available.

While I’m adding more, there’s a quote by Wittgenstein I feel is relevant here:

He (Ludwig Wittgenstein) once asked me: ‘Why do people say it is more logical to think that the sun turns around the Earth than Earth rotating around its own axis?’ I answered: ‘I think because it seems as if the sun turns around the Earth.’ ‘Good,’ he said, ‘but how would it have been if it had seemed as if the Earth rotates around its own axis then?’

I feel this has some relevance to the perceptual ruler argument. There’s a difference, I feel, between accurately perceiving the world and generating some feeling about those perceptions. While the shrunken children might feel that the dog is now tremendous, relative to their current size, they would also still be accurately perceiving the size of the dog. Indeed, putting on a heavy backpack might might a mile-long run feel more difficult, while not altering our visual perception of the distance itself.

Unfortunately, trying to test such things becomes complicated, since you’re asking the part of the brain that talks about what the part of the brain that perceives thinks. The problems are readily apparent, in that were you to ask people whether they thought pornography was “real”, they will most always tell you no while becoming physically aroused, indicating that some part of the brain disagrees with that verbal assessment.

Also, just in case you’re not already familiar with them, Rob Kurzban has a series of good posts on this topic, which can be found below:

http://www.epjournal.net/blog/2010/11/jack-jill-and-a-lion/
http://www.epjournal.net/blog/2011/11/advantages-of-error/
http://www.epjournal.net/blog/2012/10/mice-managing-mistakes/

http://www.sciencedaily.com/releases/2013/08/130807204839.htm?

Thanks for sharing these very interesting thoughts!

Needless to say, I find much to agree with in your remarks. I especially liked your insight about the usefulness (or lack thereof) of perceptual distortions that are occasioned only by prior decisions to perform the very actions that are supposedly informed by the distortions. In fact, an earlier version of the “paternalistic vision” paper had an entire section devoted to (a cousin of) your objection. To flesh it out a bit, consider the case where grasping a baton (allegedly) makes objects look closer, but only if you intend to use the baton for reaching. The case in which such a tool-induced distance-compression effect would be most useful to a perceiver is the case where grasping the baton makes reachable a formerly unreachable object. So, suppose a perceiver is in such a situation, looking at an unreachable (and, let’s assume, unreachable-looking) object, and then grabs a baton. On the assumption that action-intentions trigger paternalistic perceptual effects, the object will continue to look out of reach until the perceiver intends to reach it, and only then will the object creep closer and look to be within reach. But this leaves us with a puzzle: Why would anyone form an intention to reach an unreachable-looking object in the first place? If once-unreachable objects continue to look as unreachable as ever until baton-wielding perceivers intend to reach them, then it’s not clear how paternalistic distance-compression effects would ever obtain outside of the laboratory (where subjects were explicitly instructed to reach)—because it’s unclear to begin with why a perceiver would intend to reach an object that looked unreachable! Anyway, the objection ended up on the paper’s cutting floor for various reasons, including that it only works for positive changes in ability but not negative changes. (For example, the objections fails against the famous backpack manipulation, because the hill might look climbable initially, so you *would* intend to climb it, but then decide against it once it looked steeper.)

I very much liked your other points as well, although I’d think supporters of embodied/paternalistic vision would have some natural responses both to them and to the main themes of your post. In particular, I’d think they would simply reject the notion that there is an undistorted representation somewhere in the mind that gets contaminated by other influences to create a percept of, say, a greater distance or a steeper hill. According to (at least my reading of) the “scaling” view developed in Proffitt’s reply in Perspectives and also in a very interesting 2013 book chapter by Proffitt & Linkenauger (I can send a copy by e-mail if you’d like), the information processed by “perceptual rulers” is raw, uninterpreted visual information (e.g. retinal projections, binocular disparities, etc.), which is directly transformed into representations of spatial layout that are scaled by the relevant action capability. So, according to that theory, there’s no intermediate step wherein the visual system figures out how far some object *really* is and only then plays around with that representation in accordance with how much weight is on your shoulders; it’s that the visual system goes straight from the angular projection of that extent to its specification in units of effort, just as it does for eye-height scaling. (After all, it doesn’t seem right to say that the miniature children in Honey I Shrunk the Kids were *misperceiving* the family dog when the dog appeared huge and monster-like to them.)

For a somewhat similar reason, I’d also think supporters of the “perceptual ruler” theory of embodied/paternalistic vision would take issue with your objection to Proffitt’s argument about those rulers’ consequences for perceptual phenomenology. At least as I understand the theory, it’s not quite as you said that rulers that were merely of different sizes (e.g. a foot-long ruler vs. a yard-long ruler) would return different measurements of an extent. (That, as you pointed out, seems obviously false.) It’s more like if you had a foot-long ruler made of putty, and then, before measuring, you stretched out the ruler a bit, so that the distance between its tick-marks was increased. In that case, the ruler really would return a different measurement (in feet) for distance. But, according to Proffitt, that wouldn’t mean that the newly-stretched ruler is telling you that the object whose distance it is measuring is in a different location now than before; instead, the ruler would be telling you that that location is a different distance away now than it was before, in the native units of the ruler.

Anyway, whether or not that’s right, I think that reply fails for independent reasons (which I’ll explore in a reply of my own that I’m nearly done and am hoping to post soon). But your blog post clearly raises a number of very interesting issues that I should probably be thinking harder about!

All the best,
Chaz

The physical world (the non-living parts of it, anyway), however, is not trying to deceive us, nor can we deceive it in turn. While our perception of the world might not be objective or entirely accurate at all times, then, we should expect it to be honest. It’s our best guess, based on the information we have.

By false beliefs here, I meant dishonest or strategic ones. These are beliefs based on that best information plus something else. If, for instance, our best estimate of the width of a cliff is approximately 15m, it does us no good to then modify that guess down perceptually in the hopes that we can then jump it. A decision maker that uses this ‘true’ information and calculates the expected value of jumping will do better on the whole than a decision maker who uses ‘false’ information to calculate that expected value.

“The ultimate goal of the visual system is clearly not to precisely measure the size of an image projected onto the retina. A more behaviorally critical property of an object is its size relative to the environment, which helps determine its identity and how one should interact with the object” (Murray, Boyaci, & Kersten 2006, p. 432).

-Gary Lupyan
Assistant Professor of Psychology
University of Wisconsin-Madison

There might be some cases where an organism is designed to get things wrong (or at least fail to represent them accurately), but doing so would need to offer some compensating benefit, like helping one persuade others about something, Even then, the scope of how wrong can be is important.

I might think of examples of brain illusions that help us survive:
• We perceive the things around us as solid, but in fact there is more space than matter in the atoms. The illusion of solidity helps us in many ways, like avoiding head wounds from bumping into walls.
• We perceive the things as objects, not as atoms accidentally in positions next to each other. This enables us to manipulate objects physically and computationally. For instance if my brain presents me a group of atoms as a stone, I don’t have to calculate what every atom will do if I take the stone and throw it to a predator. Computationally it is much easier for the brain to treat that group of atoms as one object that will stay as one object when I manipulate it. But there are no objects in the nature, just atoms (or even smaller elementary particles). Despite that my mind creates the false illusion that there exist objects.