Cognitive Daily

The hollow-face illusion is one of the most dramatic and robust illusions I’ve ever come across. It’s been known for well over 200 years, but it never ceases to amaze me, as this video demonstrates:

A three-dimensional hollow face mask held a few feet away will appear to be convex (turned “out” towards the viewer) no matter which side you look at (this image is from the Max-Planck-Institut für biologische Kybernetik in Tübingen). While the movie depicts a computer-generated model, the effect works just as well with a real physical mask. Scientists have attempted to explain the illusion for centuries, but there is still much we don’t know about how it works. Our visual system can use tools like binocular disparity and motion parallax to judge distance, but these techniques don’t seem to work with the hollow mask until we’re extremely close to it: for many people, nearer than three feet. The effect is diminished if the mask is turned upside-down, but it doesn’t disappear; nearly everyone still sees the illusion.

The effect isn’t completely due to the direction of lighting either. While the visual system tends to assume light is coming from overhead, hollow masks lit from below still appear convex. Others have suggested that the illusion arises because we “know” that we’re seeing a face, and that knowledge trumps other visual cues that suggest it’s not convex like a real face would be.

If this is true, then the effect should be stronger for more familiar objects, and weaker for less familiar things. Harold Hill and Alan Johnston showed 12 volunteers three different hollow shapes: molds in the form of a teddy bear, a pineapple, and a “jelly mold” (Americans would call this a “jello mold”):

The viewers walked slowly towards each mold (facing away from them) until they could clearly see it “switch” from convex to concave, thus establishing how close they could be and still see the illusion. Here are the results:

As objects became more familiar (and arguably more human-like), people could stand closer and still see the illusion. For both teddy bears and pineapples, the illusion was stronger when they were upright, but for the jello mold, orientation made no difference. The experiment was repeated with a human face, at four different orientations. The upright face had an even stronger effect than the teddy bear, but the illusion was still present when the face was upside-down, and just as strong as the teddy bear.

Next Hill and Johnston moved to computer-generated images of faces. These were displayed stereoscopically, and viewers wore 3-D glasses. This time they systematically modified the faces, gradually adding noise. Here are a few examples:

Fourteen viewers gave each image a convexity rating, ranging from 6 (definitely convex) to 1 (definitely concave). Remember, all the images were rendered to be concave — the cues the 3-D glasses gave them suggested that these were not real faces, but hollow shells. Here are the results:

As more noise was added, making the face look less “real,” viewers were less likely to fall for the illusion, rating it significantly lower on the convexity scale. The illusion persisted longer for color faces than those rendered in gray-scale, again suggesting that the idea that we’re seeing a “real” face makes us more likely to see the face popping out towards us.

All this adds up to a fairly convincing argument that our perception of a face as a whole is what causes us to see the mask as convex, like a real face instead of a hollow shell. Our visual system is receiving a variety of different cues to depth of objects, and prioritizes them in ways that are usually quite accurate. But illusions such as the hollow face demonstrate that those priorities don’t always work. Fortunately we don’t see hollow masks nearly as often as real faces, so for the vast majority of our visual experience, our visual world seems just fine. These anomalies — what we see as illusions — can offer a powerful window into how our visual system actually works.

Hill, H., & Johnston, A. (2007). The hollow-face illusion: Object-specific knowledge, general assumptions or properties of the stimulus? Perception, 36 (2), 199-223 DOI: 10.1068/p5523

Comments

#1 phisrow
July 14, 2009

Most curious. Has anybody tried the hollow-face illusion on a selection of prosopagnosia cases?
#2 Dave Munger
July 14, 2009

Interesting thought. I didn’t know for sure, but I did some Googling around and it seems they have. It looks like there are at least a few cases of people with prosopagnosia (“face blindness” or the inability to recognize faces) who do not experience the illusion — maybe one of our readers knows more about this.
#3 KooperG
July 14, 2009

Nice article!
#4 Frank Belloni
July 14, 2009

Very neat!
#5 The Science Pundit
July 14, 2009

I wonder if this is somehow related to why we can recognize faces so much easier than AI’s can. In other words, as soon as we see enough “face like” features, or brain builds a convex 3-D model which can then be compared with the ones we have in our memory. Thus we can easily identify a face even if we’re seeing it from a completely new angle, distance, or lighting–something AI’s (so far) have trouble doing.

Of course this ability would also make us more prone to the hollow-face illusion as well as pareidolia.
#6 Peter
July 14, 2009

In the instances I’ve seen this illusion the object is made of a glossy material. The reflection of the light appears to be identical on both concave and convex sides. Could the uniformity of the light on a glossy surface, as the case with the computer generated and jello molds be tested (or is that the point and I need to do more reading). What if the object’s surface is matte, for instance cloth or paper?
#7 Charlotte
July 14, 2009

Peter: there’s a nice example you can download and make at http://www.grand-illusions.com/opticalillusions/three_dragons/ I have several printed on matte paper which work, although they work less well if there’s strong overhead lighting putting the hollow bits into shadow. Not tried them on glossy paper though.

The toy shop there also sells a solid mask version of this illusion, and various others, and there are video demonstrations of some of them. Lots of fun.
#8 Janne
July 15, 2009

It would be interesting to see if you could retrain or habituate volunteers to hollow faces. Have people live with hollow-face masks strewn about at home, by their work desk and so on for a few weeks or a month, then retest them. Remove all masks, then retest them once again after, oh, three months or so. See if the visual system is able to incorporate thing changed experience over time.
#9 Steve
July 15, 2009

I’d like to know what would happen if you asked particpants to reach out and touch the nose of the mask. Would they get it wrong? Milner and Goodale’s work would suggest that they wouldn’t. In fact, this would be an interesting test of their work.

The problem of the illusion becoming weaker when you are within arms reach is reduced with smaller faces. I have a mini version of the mask illusion sitting on my desk right now, and it is compelling at less than a foot. Unfortunately, the face is enclosed in perspex, so I can’t touch it.

These smaller faces should still allow enough room to decide if the dorsal stream is falling for the illusion. The import of this study would be to test the effect of things like face knowledge on the dorsal stream, which (when I did my MSc in 2004) was believed to work using direct perception – faces should have no effect on action, only self report.
#10 Henning Boje
July 17, 2009

Fascinating results. But doesn’t the illusion have to do with both (a) our inclination to see something-as-a-face when, at the same time, we are aware that it isn’t a face nor a representation of a face, and (b) the integration that takes place in vision which is largely “cognitively impenetrable”?
About “seeing faces”: I have no reference for this, but judging from anecdotal and private experience, I think children are notoriously eager to see-as-a-face surfaces and configurations, both natural (clouds) and artifacts (fronts and backs of cars).
About the integration in vision: it is striking that schizophrenics are resistant to the hollow-face illusion: “16 control volunteers perceived the hollow mask as a normal face – mis-categorising the illusion faces 99 percent of the time. By contrast, all 13 patients with schizophrenia could routinely distinguish between hollow and normal faces, with an average of only six percent mis-categorisation errors for illusion faces.”
(http://www.sciencedaily.com/releases/2009/04/090406102557.htm)
#11 suta
July 21, 2009

I Learn this Hollow face illusion just from saw them in Internet…and so i try it …and Now already Done My Hollow face illusion Budha…:)
#12 JL
August 6, 2009

i have face-blindness and I had to pause the video repeatedly to understand the illusion at all. Also I had to read the info to understand what illusion I was supposed to be seeing. It didn’t “work” on me the first two times I watched the video. I found myself wondering what the illusion is supposed to be!
#13 Julian Sammy
August 7, 2009

It’s the last sentence of the article that impresses me: “These anomalies — what we see as illusions — can offer a powerful window into how our visual system actually works.” In my work I constantly seek out the places where human behaviour diverges from our beliefs about our behaviour, and try to find ways to help businesses avoid the problems this can cause. Illusions like this one are a powerful way to illustrate the importance of questioning our assumptions and exposing what is, instead of what we expect or believe.