Visual cliff hanger

Vimeo has some video of what looks like footage from Gibson and Walk’s original 1960 ‘visual cliff’ experiment where they tested whether infants had depth perception by attempting to get them to walk over glass plates suspended above a drop.

Unfortunately, the video doesn’t fully describe the experiment, which is a pity as it was a fantastic idea.

The researchers wanted to find out whether 6 to 14 month-old infants could perceive depth. Babies are not the best conversationalists, but they do have a natural sense of danger, so the experiment is based on the idea that the babies will avoid perceived danger, even if it’s completely safe.

The study put the infants, one at a time, in the middle of a table, with one side replaced by glass so you could see the ‘drop’.

Their mothers would try and tempt them over both sides, and if the kids had no depth perception, the glass ‘drop’ wouldn’t seem scary and they’d just walk straight over. Those who could see the ‘drop’ would avoid it.

Pretty much none of the infants wanted to walk across the ‘visual cliff’, suggesting that even kids of 6 months old could perceive depth.

Children younger than that generally can’t crawl though, so it makes it a bit harder finding out at what age depth perception develops.

In 1973, a study by psychologist Andrew Schwartz placed five and nine-month olds on each side of the ‘visual cliff’ and measured their heart rate.

When placed over the glass ‘drop’, the five month olds typically showed no increase in heart rate, suggesting there was no danger response. This suggests depth perception probably kicks in between about five and six months old.

More recent research has shown it’s a more complex picture than this, as depth perception has many parts which don’t all seem to develop at the same rate, but the ‘visual cliff’ experiment is still widely used in psychology.

Link to video of ‘visual cliff’ experiment.
Link to text of original study.

Best visual illusion of the year announced

Mixing Memory has alerted me to the fact that the winner of the Best Visual Illusion of the Year Contest has been announced, and what a fantastic illusion it is.

It’s an animated one, so you need to go to the page and stare at the dot in the centre for 20-30 seconds.

The creators of the winning illusion, psychologists Rob van Lier and Mark Vergeer, have put up a pdf with their explanation of the effect.

And if you’re still illusion hungry after that, you can check out the rest of the finalists that came in the top 10.

On a related note, Scientific American have recently released one of their ‘special editions’ that collects V.R Ramachandran and Diane Rogers Ramachandran’s monthly articles on illusions into one magazine. I got mine from a newsagent but you can also purchase it as a DRM-free pdf online for $4.95.

Link to Top 10 2008 contest winners.
Link to Mixing Memory’s take on the winner.

A blind man hallucinating

NPR has an brief but interesting piece on a blind man who has visual hallucinations.

Stewart, the person in question, lost his sight due to hereditary sight-loss, but has developed Charles Bonnet syndrome, a curious condition where playful visual hallucinations are common.

Two things about this condition are striking: firstly, the hallucinations are typically complex and intricate but the damage is typically only to the retina, the cortex remains intact.

Secondly, unlike many other conditions where hallucinations are common, the person typically retains complete insight. They know they are hallucinating and typically don’t mistake hallucinations for the real world.

While the person interviewed in this radio segment is blind, Charles Bonnet syndrome can occur in people with partial sight, who may have only lost vision in one part of their visual field (often due to macular degeneration). In these cases, even when the hallucinations can ‘blend in’ with true vision, the person usually knows the difference.

One of the most remarkable things about the interview is that the Stewart’s hallucinations can be triggered by quite idiosyncratic things (such as foods and thoughts) and that he takes such joy in the experience.

If you want to read more about the syndrome, the Fortean Times published a great article on it back in 2004.

Link to NPR segment on Charles Bonnet syndrome.
Link to FT article on the same.

Polanski and the Professor

It was 1970, and a white Rolls Royce was gliding through the streets of London. Inside were the obviously disturbed Roman Polanski, the film director still reeling from the murder of his wife, and Richard Gregory, the legendary cognitive psychologist.

Polanski had discovered Gregory’s work on visual perception through his book Eye and Brain and decided he wanted to enlist Gregory’s help to create a 3D horror movie.

The movie was intended to be revolutionary, taking advantage of the brain’s perceptual quirks to make a truly disturbing visual experience.

They spent the week in Polanski’s office, actually the rear of his white Rolls Royce, discussing concepts, checking out studios and making plans.

In the end, their plans were too ambitious and were abandoned by Polanksi, who moved on to other projects.

Gregory remembers the episode well however, and discusses his meeting with Polanski, and the science behind their abandoned project, in an online audio recording.

Interestingly, Gregory also mentions that Polanksi also wanted to use the techniques they developed to make a 3D erotic movie.

Visual perception lectures would have never been the same again, much to the delight of generations of psychology students, but sadly, it remains only as a wonderful tale of an unlikely pairing.

The recording seems to be from a fantastic Polanski DVD box set that also contains his film Repulsion, notable for its portrayal of a young woman’s descent into a terrifying psychosis and the film’s use of perceptual distortion to communicate the experience to the viewer.

Link to audio of Gregory discussing his collaboration with Polanski.

Mind snacks

Exploratorium has a gallery of try-it-yourself perception experiments. There’s plenty of great material here, not least because of the the slightly bizarre photos of people with distracting 80s haircuts.

There are quick projects on everything from proprioception to taste, and you can tell which are the good ones because they list ‘adult help’ as one of the materials.

Think of it as the Mind Hacks that time forgot.

Link to groovy gallery of Exploratorium perception ‘snacks’.

The beauty of false depth

The image is one of many beautiful street art images from artist and architect Kurt Wenner who uses false perspective to give the images an impression of a 3D structure when viewed from a certain angle.

Wenner uses the same optical manipulation as Julian Beever, whose work we covered previously on Mind Hacks.

It takes advantage of the fact that we use visual features such as relative sizes to infer the depths of objects in the visual field.

When this is manipulated, we can be fooled into thinking that a depth is present in spatial dimensions where it can’t possibly exist – like in this case, where it seems as if the paintings represent ‘holes’ in the floor.

However, because in reality, these are flat images, the effect is lost when viewed from an alternative angle.

There are many more stunning images on Wenner’s website.

Link to Kurt Wenner’s street art portfolio (thanks Ceny!).
Link to previous post on Julian Beever’s optical street art.

Strobing numbers show saccadic vision

This week’s New Scientist has a brief letter which describes an elegant demonstration of visual processing during eye movements.

When you move your eyes (known as a saccade), visual input is suppressed, so less information is processed by the brain during the move.

This can be easily demonstrated, as described in one of the hacks in the Mind Hacks book (pdf).

An earlier article in New Scientist suggested that visual perception shuts down completely during the move, and someone wrote in with an elegant demonstration to show that this isn’t the case.

It is not strictly true that your visual perception mechanism shuts down completely during a rapid eye movement (22 September, p 34). This can easily be confirmed by flicking your eyes across a digital clock running off an alternating-current power supply, whose figures are luminous and flash at 100 or 120 hertz. You may see a line of images of the numbers, spaced out in proportion to the speed of your eye movement.

The same applies to a TV image: flicking your eyes to the right or left produces a succession of lozenge-shaped images, whereas flicking up or down results in a series of images respectively drawn out or squashed. If you flick your eyes down fast enough you can reduce the picture to a single bar, and if you flick your head down at the same time you can even manage to invert the picture, though you have to be very quick. Do not attempt this, however, if anyone is watching you.

Link to NewSci letter ‘Saccade effects’.
pdf of hack to demonstrate visual suppression during saccade.

Artificial intelligence ‘sees’ visual illusion

A study just published in PLoS Computational Biology has reported that an artificial intelligence system trained to make sense of a simulated natural environment is susceptible to some of the same visual illusions that humans fall for.

In one of these, the ‘Herman grid‘ illusion – illustrated on the right, you may be able to ‘see’ fuzzy patches of grey in the white stripes, despite the fact that there is no grey in the image (click for a bigger version if it’s not clear).

David Corney and Beau Lotto, researchers working in the Lotto Lab (which has a wonderful website by the way), have been training artificial intelligence systems to distinguish surfaces in a simulated natural environment with lots of ‘dead leaf’-like shapes.

When training these sorts of systems, the idea is not to program them with specific rules, but to present an image and let the neural network make a guess.

The researchers then ‘tell’ the AI system whether it is correct in its guess, and it adjusts itself to try and reduce the extent of the error on the next guess. After many learning trials, these sorts of ‘back propagation‘ neural networks can make distinctions between quite complex stimuli.

In this case, Corney and Lotto decided that once the system was fully trained to complete its task successfully, they would test it with some visual illusions experienced by humans.

Interestingly, the AI system was susceptible to the Herman Grid illusion, sensing ‘grey’ where there was none. Other illusions produced similar results.

The fact that both humans and AI system ‘fall’ for the same illusions, suggests that they take advantage of visual abilities that have been shaped by our experience of the visual world.

Link to paper in PLoS Computational Biology (thanks Matt!).
Link to study write-up from the university’s news site.
Link to Lotto Lab website (with loads of cool images and demos).

Illusory motion with waves of almonds

I’ve just found a visual illusion that gives a striking impression of motion from a static image. It’s entitled ‘this picture is not animated’, which, like anything eye-catching on the internet, immediately made me check whether it was or not.

With many of these sorts of illusory motion images, you can ‘stop’ the motion by simplying viewing them through a very small aperture.

Putting a pin through a piece of paper and viewing it through the hole does the trick, but so does making a small viewing hole with your fingers.

As you can see yourself, the picture stops ‘moving’ when viewed like this, but starts again as soon as you view it normally.

This also prevents stars from twinkling when you view them at night. The traditional explanation of ‘star twinkle’ is that the light gets bounced around as it travels through the atmosphere, giving it the twinkling effect.

In fact, by looking at them through a small hole, you’re preventing any effects caused by your eyes moving about.

The fact that the illusion stops moving and the stars stop twinkling when you do this, suggests that the way our eyes scan across the visual scene is an important part of why we see the false movement in these sorts of images.

Because of this, you can ‘speed up’ and ‘slow down’ the false movement in the visual illusions by changing how often you move your eyes.

UPDATE: Thanks to celeriac for posting a link to a scientific paper which explains this effect.

Link to striking movement illusion.

Pink slip, feeling blue

Ben Goldacre over at Bad Science has written a great analysis of a recent study that suggested we have the traditional ‘pink for girls, blue for boys’ because of evolutionary differences in colour preference.

However, it seems not only are the study’s findings not strong enough to make an evolutionary claim, but that the ‘pink for girls, blue for boys’ idea is relatively recent and hardly as traditional as we like to think.

The data itself is interesting if not a little unspectacular. Men and women from the UK showed different colour preference curves with men showing a preference for bluer shades over women.

In a sample of Chinese participants the preference was much less pronounced and peaked at more redder shades overall.

One of the curses of evolutionary psychology, the science that attempts to work out whether any of our psychological preferences are the result of natural or sexual selection, is that any sex difference is fodder for an evolutionary explanation.

Actually, we know there are definite differences in colour perception between men and women. There’s a great paper that summarises the scientific evidence which available online as a pdf.

There are sex-linked differences in specific genes that are linked to colour perception, which is why men are more likely to be colour blind and perhaps 1% of women may have four, rather than three, colour receptors in the retina.

But as Ben points out, simply finding a sex difference in colour preference really doesn’t tell us anything about genetics or evolution. It could easily just be an effect of culture or fashion.

Link to Bad Science on pink-blue study.
pdf of paper on genetics, sex differences and colour perception.

Spinning silhouette illusion

I’ve just found this ‘spinning silhouette‘ visual illusion which took ages to take effect but when it did it was so striking I thought at first it was faked.

The idea is that you keep looking and the woman suddenly ‘flips’ and seems to spin in the opposite direction. It’s very impressive when it happens, but it seemed to happen so randomly that I wondered whether it had been programmed to randomly reverse.

However, I’ve found that if you cover image apart from the shadow of the feet and concentrate on seeing them rotate in the opposite direction, when you uncover the image, it too will seem to be in a reverse spin.

I’m guessing it works because our brain is making the best guess of a 3D shape from a 2D image. The silhouette from a real 3D rotating shape would look identical no matter what way it rotated.

Think about a rotating coin. No matter which way it turns, the silhouette would be the same – it would seem as if a disc was being progressively ‘squashed’ into a line and then back to a disc again.

As with all visual perception, our brain ‘fills in the gaps’ with best guesses, in this case to make it seem like a rotating 3D shape.

However, there’s actually no information about which way its rotating, so it can suddenly ‘flip’ when our perception of the direction becomes unstable and another interpretation takes effect.

It’s like a motion-based necker cube effect.

Link to Spinning Silhouette illusion.

An artistic impression of alcoholic delirium

The picture is from this month’s British Journal of Psychiatry and is entitled ‘Memory image of acute alcoholic delirium’.

It was included in a 1919 book of cases studies of people with alcoholic delirium, otherwise known as delirium tremens or the DTs, and was drawn by a patient to communicate their hallucinatory experiences.

Delirium is a mental state where hallucinations and delusions are present, but unlike psychosis, there are also severe impairments in consciousness and cognitive function.

It typically resolves quickly, usually when the physical disturbance that caused it (e.g. fever, intoxication) subsides.

The author of the book, the Danish psychiatrist Einar Brünniche, explains the image:

‘Finally, I should like to present an image, a reproduction of a coloured drawing, in which a patient, an artist, without words, but none the less very effectively and vividly, describes the memory of his past, alcoholic delirium… It shows us the many facets of hallucinations, their animal imagery, their life and mobility and their partial transformation of real objects; it shows us the air brimming with cobwebs, threads and smoke.

However, I should think that the image illustrates a stage at which the delirium has not yet reached its zenith since the patient is still bedridden. True, the hallucinations seem spooky, but they have not yet filled him with uncontrollable dread; he has not yet been stirred to action, he has not yet taken steps to ward off the danger. Besides, the picture speaks for itself’.

There’s more at in this brief ‘psychiatry in pictures’ article at the link below.

Link to British Journal of Psychiatry full image and article.

Striking perspective shift illusion

I’ve just stumbled across a remarkably simple yet fiendishly effective visual illusion that seems to give flat images the illusions of 3D depth. It works by quickly shifting between two images of the same scene taken from slightly different perspectives.

As we’ve noted when discussing other illusions, our brain generates the experience of seeing a 3D world by making the best it can out of the two fairly poor quality flat images that fall on the back of the retina.

Visual illusions usually have their effect by taking advantage of the brain’s processes for inferring visual features.

These processes are really nothing more than educated guesses, and illusions essentially give the brain red herrings, misleading it to guess in the wrong direction, so we experience one visual feature in a context where it wouldn’t normally appear.

In this case, the illusions give the misleading impression of depth in the context of an entirely flat image.

One clue the brain uses to infer depth is occlusion – things that are near the front of a visual scene will block those at the back. They just get in the way.

In fact, the lines in cartoons or diagrams are often just a representation of the occlusion contours – the edges of where one surface hide another.

However, although we can see that a cartoon drawing of a head is a 3D representation, we don’t experience it as actually having depth. It doesn’t seem to stand out from the page.

The perspective shift illusion flips between two images taken from slightly different perspectives and this seems to add a dynamic aspect – the illusion of movement.

Movement is often crucial for determing something’s depth. Have you ever moved your head side-to-side when you see a puzzling sculpture to try and better understand its shape?

This allows us to see the relative depth of each of the occlusion contours by experiencing how much the foreground moves in relation to the background. Things nearer the front seem to ‘move quicker’ – something known as motion parallax.

I’m guessing by adding an impression of movement – some motion parallax – to a detailed photograph that already has other depth cues from the natural environment, the perspective shift illusion produces an impression of real depth.

If you want to know more about the cognitive science of 3D shape perception, there’s a great review article by available Dr James Todd available online as a pdf file.

Otherwise, just spend a few minutes checking out the impressive visual effect.

Link to perspective shift visual illusion.

Single gene gives mice new sense of colour

The journal Science reports a study showing that mice given a single gene can develop full colour vision. Mice, like most mammals except primates, are normally colourblind. The implanted gene, which is found in humans, is responsible for making a photopigment, a light-sensitive protein in the photoreceptors of the eye. The researchers from the Howard Hughes Medical Institute validated their findings with cell recording and with behavioural tests, demonstrating pretty conclusively that the mice really can see in colour, being able to make discriminations normal mice cannot, and this is because their photoreceptors are sensitive to long wavelength red light.

Because only a single new gene has this effect, the study is reported as demonstrating that primate colour vision could have evolved very suddenly. However, this angle is perhaps less suprising if we consider that colour vision is phylogenetically ancient – primate colour vision doesn’t represent the first time it has evolved, rather primate colour vision is more of a recovery of the function which is found in many non-mammal species such as reptiles. The structural correspondence of this is that the appropriate apparatus for colour vision is extant in mammals – it is just that non-primate mammals lack the appropriate variety in their photopigments. The study is is another demonstration of the amazing ability of the brain to adapt to and take advantage of whatever sensory input is available to it (related to this, see this article on human tetrachromacy, via Slashdot)

Movies and the McGurk Effect

HacksZine is hosting a video by Brian Sawyer who riffs on the Mind Hacks book entry on the McGurk Effect and shows how this is used in movies.

The McGurk effect is, for example, where when you hear the sound of someone saying ‘Ba’ at the same time as you see them saying the sound ‘Ga’, you hear the second, because the information from your vision shapes how you perceive the sound.

Sawyer notes that this is commonly occurs in movies when they’re dubbed, so despite the character saying ‘You lousy melon farmer!’ when this was obviously not what was said in the original, the dubbing doesn’t seem completely out of whack.

Link to ‘Hear with Your Eyes: The McGurk Effect’ from HacksZine.